RETINAL VEIN OBSTRUCTION* vein, gradient optic keepingretinal venous obstruction) is dueto arterial narrowing or occlusion, rather than to the effects ofhaemorrhage." Foster Moore(1924)

Brit. J. Ophthal. (1962) 46, 336.

RETINAL VEIN OBSTRUCTION*BY

SIMON BEHRMANMoorfields Eye Hospital

AT the root of the clinical phenomena to be described, lies a singularhaemodynamic arrangement to be found in the central retinal vein. Withina space of a.few millimetres, the pressure in this vessel drops from around20 mm. Hg at the optic disc, to levels ranging from zero in the uprightposition, to 10 mm. in recumbency. The examination of the vein, however,reveals no special device whereby this high-pressure gradient is maintained.After entering the nerve-head, the vein traverses the lamina cribrosa and thenmaintains its central position within the optic nerve for a further distance ofa'few millimetres. The lamina cribrosa, being able to withstand these pres-sure differences, is the only structure in the course of the central retinal veincapable of acting as a"throttle" device to maintain this pressure gradient.Being unyielding within this pressure range, the interstices of the laminacannot be anything but fixed in size. Poiseuille's formula would suggestthat, in order to bring about a 50 per cent. reduction ofpressure after emissionthrough the lamina cribrosa, the diameter ofthe vein would have to be reducedby only some 16 per cent. In keeping with this inferred "throttle" actionof the lamina cribrosa are the observations that, in cats and dogs, pressurevariations in the perioptic subarachnoid space below 18 mm. Hg failed toaffect the pressure of the central retinal vein at the nerve-head (Gibbs, 1936).The conclusion that the pressure gradient within the retinal vein is maintainedby a high "flow resistance" is inescapable, though for the purpose of thethesis to be presented, its precise topology is immaterial.Two clinical phenomena occasionally encountered in association with

obstruction of the central retinal vein must now be considered:

(A) PRESERVATION OF GooD VISION.-The following cases illustrate sparing ofvision despite ophthalmoscopic evidence of severe embarrassment of the retinalvenous circulation.

Case 1 (Mr. Greaves).-When the optician supplied this patient with glasses, he indicatedhis wish to see her again after one year. In compliance with this request, she called after

* Received for publication July 12, 1961.336

on May 14, 2021 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.46.6.336 on 1 June 1962. D

ownloaded from

http://bjo.bmj.com/

RETINAL VEIN OBSTRUCTION

this interval, and he was surprised to find evidence of central retinal venous obstruction(Fig. 1). When questioned she admitted that she had recently become aware of a triflingdiminution of vision in the left eye, and this was found to be reduced to 6/12.

FIG. 1.-Central retinal venousobstruction revealed by fundusphotograph in Case 1.

This case thus confirms that extensive extravasation can occur in retinal venousobstruction with no catastrophic damage to the retinal function. A profuseretinal haemorrhage, especially in the region of the macula, will lead to an -appreci-able lowering of vision, but Case 1 suggests that, after resolution of the haemor-rhage, the visual acuity may be expected to return to normal were the haemorrhagethe sole reason for loss of vision. A retinal haemorrhage in retinal venous"thrombosis" can disappear and "leave no discoverable residue" (Foster Moore,1924).

Case 2 (Mr. Leigh).-On July 11, 1961, "mistiness" appeared suddenly in a part of thelower temporal quadrant of the right field of vision. Gradually this mist condensed intoan opacity "like a black line" which totally obliterated vision in this part of the field. Ina space of a few hours vision returned to normal, but the scotoma recurred for severalhours on each of the following few days. When the fundus photograph (Fig. 2) wastaken on July 18, the uncorrected visual acuity in the right eye was 6/6 (ptly.).

FIG. 2.-Fundus appearance oneweek after first examination inCase 2.

23

337



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

SIMON BEHRMAN

(B) AMAuRosIs FuGAx.-Sudden, brief episodes of obscuration of vision of theaffected eye, ranging from slight "veiling" to complete "blindness", is not anuncommon symptom preceding established venous obstruction.

Cam 3 (Mr. Leigh).-Recurrent transient obscurations of the vision of the left eye for2 months were followed by persistent amblyopia attributable to retinal venous obstruction.

Case 4 (Mr. H. Ridley).-For the last two years there had been brief episodes ofrepeated severe obscuration of vision of the eye, which was now the seat of central retinalvein obstruction. Examination was negative, except for moderate hypertension.

Among eighteen cases of amaurosis fugax briefly reported by Uhthoff (1925),there was an unspecified number of cases of "retinal venous thrombosis".Similar examples can be found among clinical histories cited by Morgan (1953)and Stephenson (1956). The fleeting nature of anlaurosis and the completerecovery of vision between attacks are also features which characterize the commonform of amaurosis fugax due to retinal circulatory insufficiency caused by carotidocclusion.

It seems highly probable, therefore, -that amaurosis fugax preceding obstructionof the central retinal vein may be similarly due to transient retinal ischaemia causedby a reduction of either arterial or capillary circulation. This transient retinalischaemia in response to central retinal venous obstruction is presumably engen-dered by a local veno-vasomotor reflex which can be visualized as one of the factorswhereby retinal venous homeostasis is maintained.

Veno-vasomotor Reflex

In a carefully controlled study, Haddy and Gilbert (1956) demonstratedthat raising the systemic venous pressure in the foreleg of a pentobarbitalanaesthetized dog elicits reflex "small vessel" constriction. After procaineblock of the limb, the "small vessels" fail to constrict in response to venousobstruction. Arterioles, capillaries, and venules were regarded as the "smallvessels" in these studies, and the authors placed the resistance at the arteri-olar level. The presence of a "venous small vessel" reflex had previouslybeen suggested by Girling (1952), who used an entirely different technique.Haddy (1956) demonstrated that raising the renal venous pressure in alaparotomized dog elicited a renal vascular constriction. Since the samemanoeuvre failed to produce a significant resistance change in denervatedkidneys, he inferred the existence of a renal venous-arteriolar reflex.A similar veno-vasomotor reflex mechanism, if present in the eye, would

presumably be initiated by baroceptors so arranged in the retinal veins thatthey would be activated by changes in radius of these vessels. The magni-tude of transmural pressure, i.e. excess of intraluminal over intra-ocularpressure, would determine the change in the radii. Such a low-pressurebaroceptor mechanism of the peripheral vascular bed, while interlockedwith other factors influencing circulation as a whole, may, however, actindependently and in a contrary direction to that of other factors controlling

338



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/


the system. Kurus (1952) and Wolter (1957) demonstrated that smallretinal veins were well provided with terminal nerve plexuses, which mightbe assumed to be in continuity with the baroceptors.Two important points emerge from these studies:(i) The maximum rate of flow through the central retinal vein is probably

fixed within narrow limits;(ii) A veno-vasomotor reflex may be an important factor controlling rate

of flow through the central retinal vein.

CENTRAL RETINAL VENous OBSTRucTIoN.-This occurs when the maximumrate-of-flow capacity has been reduced, either by a rise in the viscosity of theblood, or by stenosis of the lumen of the vein.

(1) Raised Viscosity ofBlood.-Given a vascular channel of dimensions towhich Poiseuille's formula applies, the relationship-for any given pressure-between rate of flow and viscosity should be inversely linear. Because ofthe adaptability of the vascular bed, the rate of flow of systemic blood is notreadily affected by rises in viscosity. However, changes indicative of areduced flow become apparent without delay in the central retinal vein inresponse to an increasing viscosity. This limited ability of retinal venouscirculation to accommodate to rises in viscosity is singular, and is presumablydue to fixed, high "flow resistance" at the lamina cribrosa. Ophthalmo-scopic features characteristic of slowly progressive central retinal venousobstruction are found early in all those medical conditions in which anenduring elevation of blood viscosity is known to occur. These funduschanges are represented by:

(a) Tortuosity and looping of veins, at first without haemorrhages;(b) Dilatation of retinMl capillaries with aneurysm formation.

Among such diseases may be mentioned polycythaemia vera, erythro-cytosis due to emphysema, sickle-cell disease, Waldenstrom's macroglobu-linaemia, hyperproteinaemia, and diseases such as cystic fibrosis of thepancreas, in which there is erythrocytosis due to pulmonary insufficiencyand elevation of gamma globulin; the leukaemias must also be included(Bruce, Denning, and Spalter, 1960). The appearance of retinal veins,therefore, may afford early information on enduring blood "hyperviscosity".

(2) Stenosis.-The yielding and backward bulging of the lamina cribrosain glaucoma might be expected to bring about a stenosis of the central retinalvein as it leaves the eye-ball. The high flow resistance in this region willbecome even higher, and venous obstruction is likely to take place. Theonset of obstruction may be heralded by repeated episodes of amaurosisfugax.

339



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

SIMON BEHRMAN

Case 5 (Mr. Stallard).-An otherwise fit man had amblyopia ex anopsia of the right eye.About 3 weeks previously he had to stop driving his car repeatedly whenever his left eyesuddenly became blind, the condition persisting for a minute or so. These episodesincreased in duration, and after recurring about half-a-dozen times in the course of theday, vision returned to a very limited extent only. He was found to have bilateral chronicsimple glaucoma with obstruction of the left central retinal vein.

Verhoeff (1913) was among the first to consider the effect of "traction pro-duced by a receding lamina cribrosa on the central retinal vessels". A cleardistinction between "thrombotic" ("haemorrhagic") glaucoma and pre-existing glaucoma with central venous obstruction as a complication wasmade by Redmond Smith (1955), who also drew attention to the frequencyof this complication. In differentiating this condition from "thrombotic"glaucoma, he relied on the presence of increased tension in both eyes, and onthe absence of rubeosis iridis.Vannas and Tarkkanen (1960) claim to have been able to demonstrate the

presence of pre-existing simple glaucoma in 42 per cent. of their cases ofcentral retinal vein occlusion. No exception can be taken to their statementthat "the discovery of central retinal vein occlusion calls for exhaustiveexamination for primary glaucoma". The relatively high proportion ofpre-existing glaucoma among their cases of venous occlusion may, however,be misleading, since it was based mainly on tonography.Dobree (1957) drew attention to the occasional coiling and looping of the

vein at the disc showing cupping. It is unlikely that this localized morpho-logical abnormality of the vein is indicative of venous obstruction. Wybar(1957) suggested (more probably) that the looping was congenital and ren-dered visible by cavitation. The normal lamina cribrosa "varies consider-ably in toughness and resistance", and the degree of backward bowing of thelamina cribrosa in glaucoma is subject to variation (Samuels and Fuchs,1952). This may account for the absence of retinal venous obstruction inmany cases of glaucoma.Many other pathological conditions, among which atherosclerosis is pre-

eminent, may seriously embarrass the retinal venous circulation by causingsome stenosis of the vein in the same situation as in glaucoma (Morgan,1955).

VENO-VASOMOTOR REFLEX UNDER PATHOLOGICAL CONDITIONS.-An excessiverise in "transmural pressure" in the retinal veins under the conditions pre-vailing in obstruction might be expected to produce such a degree of retinalarteriolar constriction as to abolish temporarily retinal function. If thisischaemia is of adequate degree and duration, the retina may suffer irrever-sible damage. The intimate relationship between retinal venous and arterialcirculations has been noted before, but the explanation put forward was thatthe venous "thrombosis" was a consequence of transient arrest or of aslowing of the retinal arterial circulation (Leber, 1915). However, in the

340



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/


light of our recent knowledge of recurrent episodes of retinal circulatoryinsufficiency in cases of carotid occlusion, this explanation is untenable(Behrman, 1954). Lister and Zwink (1953), who reported several cases oftributary vein thrombosis with corresponding sector field defects, conclude:"It seems reasonable to suppose that the permanent defect in all cases (ofretinal venous obstruction) is due to arterial narrowing or occlusion, ratherthan to the effects of haemorrhage." Foster Moore (1924) had found that,in at least 42 per cent. of his cases, different degrees of loss of sight occurredin an apoplectiform fashion. This manner of loss of vision could be attri-buted to activity of the veno-vasomotor reflex.

Braendstrup (1950), in a monograph on thrombosis of the retinal veins,observed that "the variations in the visual acuity do not always reflect theophthahmoscopic picture". Further, there is general agreement with theview expressed by Lister and Zwink (1953) "that as a rule the final vision isapproximately the same as at the outset, but there are exceptions". Theseclinical observations, as also the disappointing results of anti-coagulanttreatment, could be explained on the basis of irreversible retinal damageengendered by an abnormal duration and intensity of the veno-vasomotorreflex.

Summary

(1) The pressure gradient within the retinal vein is maintained by a high"flow resistance".

(2) This, by fixing within narrow limits the maximum rate of flow throughthe retinal vein, accounts for the facility with which retinal venous obstruc-tion occurs in a variety of pathological conditions.

(3) Transient episodes of amaurosis preceding venous obstruction suggestthat a veno-vasomotor reflex may be an important factor in controlling rateof flow through the central retinal vein.

(4) Excessive activity of the veno-vasomotor reflex under pathologicalconditions probably accounts for the variable degree of loss of retinal func-tion which may follow obstruction of the central retinal vein.

(5) The earliest clinical indication of enduring blood "hyperviscosity" isprovided by central retinal venous obstruction.

(6) A recession of the lamina cribrosa in glaucoma, by raising further thehigh normal "flow resistance" in the retinal vein, often causes clinical andophthalmoscopic manifestations of retinal venous obstruction. This canbe the first clinical manifestation of glaucoma.

I should like to thank Drs. Hugh Davson and Donald A. McDonald, and Mr. E. S. Perkinsfor their great patience and helpful comments, and also the surgeons of Moorfields Eye Hospitalfor permission to publish the case histories.

341



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

342 SIMON BEHRMAN

REFERENCES

BEHRMAN, S. (1954). Postgrad. med. J., 30, 570.BRAENDsTRUP, P. (1950). Acta ophthal. (Kbh.) Suppl., 35.BRUCE, G. M., DENNNG, C. R., and SPALTER, H. F. (1960). A.M.A. Arch. Ophthal., 63, 391.DOBREE, J. H. (1957). Trans. ophthal. Soc. U.K., 77, 229.GIBBs, F. A. (1936). Arch. Neurol. (Chicago), 35, 292.GIRLING, F. (1952). Amer. J. Physiol., 171, 204.HADDY, F. J. (1956). Circulation Res., 4, 659.

and GILBERT, R. P. (1956). Ibid., 4, 25.KuRus, E. (1952). Klin. Mbl. Augenheilk., 121, 318.LEBER, T. (1915). In "Graefe-Saemisch Handbuck der gesamten Augenheilkunde", 2nd ed.,

band 7, teil 2, Kap. XA, 1 halfte, p. 355. Engleman, Leipzig.LisTER, A., and ZwiNK, F. B. (1953). Trans. ophthal. Soc. U.K., 73, 55.MooRE, R. FosTER (1924). "Retinal Venous Thrombosis". Supplement to Brit. J. Ophthal.

Pulman, London.MORGAN, 0. GAYER (1953). Trans. ophthal. Soc. U.K., 73, 70.

(i955). Ibid., 75, 3.SAMuEiLs, B., and FucHs, A. (1952). "Clinical Pathology of the Eye". Cassell, London.SmrrH, REDMOND (1955). Trans. ophthal. Soc. U.K., 75, 265.STEPHENSON, R. W. (1956). Ibid., 76, 253.UHTHOFF (1925). Klin. Mbl. Augenheilk., 75, 469.VANNAS, S., and TARKKANEN, A. (1960). Brit. J. Ophthal., 44, 583.VERHOEFF, F. H. (1913). Arch. Ophthal. (N. Y.), 42, 145.WOLTER, J. R. (1957). v. Graefes Arch. Ophthal., 158, 524.WYBAkR, K. (1957). Trans. ophthal. Soc. U.K., 77, 238.



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

Documents

RETINAL VEIN OBSTRUCTION* vein, gradient optic keepingretinal venous obstruction) is dueto arterial narrowing or occlusion, rather than to the effects ofhaemorrhage." Foster Moore(1924)